INDEX

Explanations

words related to a particular trading card game, including specific card names and game strategies

oai_token-act-pair · gpt-3.5-turbo

references to specific game mechanics or character abilities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enegger

-0.80

 certific

-0.80

internet

-0.77

 sitcom

-0.77

çīĪ

-0.76

 Sach

-0.75

Þ

-0.74

CES

-0.73

 educational

-0.73

 blogging

-0.72

POSITIVE LOGITS

blade

1.24

Blade

1.11

 Cloak

1.04

 Armor

1.02

 Cursed

1.02

 Paladin

1.01

Spell

1.01

 Warlock

1.01

 Blade

1.00

 Grimoire

1.00

Activations Density 0.193%

words related to a particular trading card game, including specific card names and game strategies

references to specific game mechanics or character abilities

No Comments

No Known Activations